1-to-1-abm-ads

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). LinkedIn ad-building workflow ingests outsider-authored free text from user-provided per-company config fields like commentary, headline, and landing_page (from --config), which are then POSTed into the LLM-created creative/ad content via scripts/build_campaign.py Phase 6 (build_ad) without any prior selection of a trusted item.